Telephony, the Internet and Technological Growth
There's been a great deal of talk about the growth of the Internet and the growth of telephony (both wireless and fixed). But a look at history (since 1876 for the telephone and since 1969 in the case of the Internet) reveals that the curves are amazingly similar. Is a useful generalization to be made about technology growth and take-up in general?
All human endeavors involve missteps. Moving ahead with technology is no exception.
“Why would anyone want a telephone in their home?” asked Alexander Graham Bell in 1878, responding to a reporter's query. Nearly a century later, in 1977, Ken Olsen, founder and president of DEC, asked, “Who would want a computer in their home?” in response to a suggestion.
I'll return to these anecdotes later. I find it interesting to look at the development of technology and its reception – Bell and Olsen weren't ignorant – though some earlier data isn't as solid as it might be.
We all know the steamship was called, “Fulton's Folly” (1807). Semmelweis was ostracized in Vienna for suggesting that physicians and midwives wash their hands. (He was committed to an asylum where he was beaten to death by the guards at age 47.)
The Early Years of the Phone
It was July 1875 when Bell uttered his famous words, “Mr. Watson, come here, I want you,” to Watson just off Cambridge Street in Boston. His patent (#174,465) is dated March 3, 1876, Bell's 29th birthday. Yet the FCC states that there were 2,600 telephones in use that first year.
A number of points are relevant here, not least of which is that two “instruments” were required for each call. Each wire ran point-to-point: within a factory; from a lumber yard to a furniture manufacturer; etc. The first “exchange” was set up in Philadelphia two years later.
But three months after the Bell patent was awarded, there was the Centennial Exposition.
In a sweltering hall, Dom Pedro, the Emperor of Brazil, picked up the receiver, listened, and announced: “My God – it talks!” His publicly recorded astonishment was followed by other notables: Joseph Henry (discoverer of electrical inductance and first Secretary of the Smithsonian) said of the invention that, “This comes nearer to overthrowing the doctrine of the conservation of energy than anything I ever saw.” Sir William Thompson (Lord Kelvin) called it, “The most wonderful thing I have seen in America.”
Fast-forward to another innovation of technology: 93 years later, in 1969, the ARPANET comprised four nodes: UCLA, the Stanford Research Institute, UC Santa Barbara and the University of Utah.
Let's compare the first 25 years’ growth of both the phone and the Internet: 1876 – 1901 and 1969 – 1994.
In 1901 there were 1.8 million phones in the US (FCC Statistics), 693 times the 1876 number. In 1994, there were 3.8 million Internet hosts worldwide, nearly a million times the four connections that existed just a quarter century earlier (estimate by John S. Quarterman). In 1880, there were 1.1 telephones per 1,000 people; in 2013 there are more telephones (land-line and mobile) than people on this planet.
The proliferation of the telephone is vital to my story: Without the phone and telephone lines and towers, there could have been no Internet. (Computers are another piece of the tale, but I'll not go into that here. J.C.R. Licklider, one of my heroes, said that computers were communication devices, not calculators, in 1968, before there was a ‘Net.)
Initial adoption isn’t necessarily the most important “trend.” The growth of telephony was not overwhelming: 2,600 to 9,300 to 26,300 to 30,900 to 47,900 in 1880. In 1897 there were 515,200 telephones in service.
In 1930, Herbert Casson began The History of the Telephone with, “Thirty-five short years, and "presto!" the newborn art of telephony is full grown. Three million telephones are now scattered abroad in foreign countries, and seven millions are massed here, in the land of its birth.”
In the first five years, ARPANET hosts numbered 4, 13, 23, 35, to 49 in October 1974. It was only in the 1980s that host numbers began to double each year: 235 in 1982; 562 in 1983; 1024 in 1984; and then 1,961, 2,308 to 5,089 in 1986 – when the growth began to explode. In July 1989 there were 130,000 hosts. Just 18 months later, there were more than half a million.
And that's what grabbed my attention: In very different circumstances, a century apart, it took just over 20 years for the telephone to reach the point where the term “usage exploded” was relevant, and for the Internet to reach over 500,000. Why this similarity and why the mushrooming growth of each over the next 20 years?
For more instruction, let's look at a totally different technology: the passenger car. While the first (steam-powered) vehicles date from the 1700s, the gasoline-powered internal combustion engine was developed by Nicklaus Otto in 1862 and the four-stroke cycle produced by him in 1876. The first sale of a motor car – manufactured by a German company (Benz) and sold to a Frenchman (M. Emile Roger of Paris) – was recorded in 1888. Duryea produced the first car in the U.S. in 1893.
On September 13, 1899, Henry Bliss stepped out of a streetcar in New York City and was struck by an electric taxi – receiving the dubious distinction of becoming the first reported automobile accident fatality in America. (In 2000, the worldwide annual toll was 330,000.)
But real growth numbers are hard to come by. It's been stated that there were 300 cars in the United States in 1895; 78,000 in 1905; 459,000 in 1910; and 1.7 million in 1914. However, this doesn't quite line up with the annual production of automobiles in the United States, which increased rapidly in the early 1900s, from less than 5,000 in 1900 to more than 1.5 million in 1916.
But we do know that Henry Ford introduced the Model T in 1908; 17,000 were sold that first year. By the time the T went out of production in 1927, more than 15 million had been sold.
So, in about 20 years, that Ford went from zero to 17,000 to 15,000,000.
Thus, the adoption curve for the automobile over two decades is steeper than that of the telephone and of the computer network.
I began this article with quotes from Bell and Olsen. Bell's invention was an accident; he was trying to develop a hearing aid (his wife, Libby, was deaf). Bell recognized the telephone for what it was, but it was still an instrument for an individual. By the way, Bell’s principal reason for not having a phone at home was that people might call at any time, possibly interrupting dinner; 125 years later, we all have cell phones so that we can be interrupted no matter where we are and so that people I hardly know can send me pictures of their dinner or their cat.
Olsen's company manufactured relatively large computers. He meant more that people wouldn't want a DEC PDP 10 (a memory module alone was 30 x 75 x 30 in.) at home. Keep in mind that the IBM PC and the Osborne (the first portable computer) appeared in 1981.
Three decades earlier, T.J. Watson's advisors at IBM predicted a global computer market of five. But they were thinking of machines like the Mark I: 51 feet (16 m) in length, eight feet (2.4 m) high, and two feet (~61 cm) deep. It had a weight of about 10,000 pounds (4500 kg). But in 1945 it was the only game in town.
Casey Stengel remarked, “Never make predictions, especially about the future.” But Bill Gates (at Comdex 1994) said, “I see little commercial potential for the Internet for at least 10 years.”
What is clear is that new technology has always begun slowly, but that the slope of the curve becomes quite steep until it flattens or takes a steep dive. The latter is usually because something else has come along. A new Apple iSomething will have a large initial sale, because it builds on a previous iSomething.
The telephone had no true predecessor, nor did the Internet. The automobile was the successor to the cart and the carriage, and so its growth was different. (And you fed it petroleum, rather than hay and oats.)
I think the truly important thing is that new technologies begin quite slowly, requiring decades to take off. New methods of doing the familiar involve far more rapid growth. Think of email: It was not even thought-of in 1969; invented by Ray Tomlinson in 1971; the largest part of net traffic by 1973 (yes, just two years later). The telephone and the Internet were revolutionary; the automobile and email were new ways of doing something familiar.
[Vain footnote: For data and information on the Internet, see P.H. Salus, Casting the Net (Addison-Wesley, 1995); on the history of computing, see P.H. Salus, A Quarter Century of UNIX (Addison-Wesley, 1994).]